Dec 2021 – Dec 2024 · 3 yrs

• Developed SageMaker Distributed Model Parallel library, which makes large scale distributed deep learning training easier and performant. Optimized multi-GPU training of LLM’s like Transformers, Llama and Mistral by leading 5-person team.

Aug 2020 – Dec 2021 · 1 yr 4 mos

• Part of Amazon Aurora machine learning ecosystem team with focus in model design for product recommendation system where customer can generate recommendations by writing plain SQL queries.
• Developed novel prefetch algorithm for data read ahead by incorporating clustering algorithms to forecast based on data access patterns which increases SQL read throughput by 30%.

Jun 2019 – Aug 2020 · 1 yr 2 mos · Greater Bengaluru Area

• Developed Genshop which is Adobe's in-house deep learning library for Generative models primarily GANs.
• Worked on GPU programming with CUDA primitives for Adobe Color Engine, Graphics engine and Adobe font engine

May 2018 – Jul 2018 · 2 mos · India

• Interned with Big Data Experience Lab, Adobe Research on the project to personalized highlighting based on reader feedback. Incorporated information retrieval and reinforcement learning techniques for keyword extraction.

Dec 2017 – Jan 2018 · 1 mo · Roorkee, Uttarkhand

• Researched on action recognition with feature fusion which involved combining traditional computer vision techniques with deep learning to improve the accuracy of action recognition.